Research On Control System Of Quad-Rotor UAV Based On Active Disturbance Rejection Controller

Compared with the traditional fixed-wing UAV,the quad-rotor UAV has better flexibility and maneuverability.For this reason,quad-rotor UAV can accomplish difficult tasks in complex environments.Now days,quad-rotor UAV had been widely used in military,police,commercial,civil and other aspects.The flight control of quad-rotor UAV was difficult because of underactuated,nonlinear,strong coupling,multivariable characteristics.Flight control technology of quad-rotor UAV was one of hotspots in current research.In this paper,the research of UAV controls system mainly included the following contents:Firstly,according to the rigid structure and flight principle of the quad-rotor UAV,a nonlinear dynamic model of the quad-rotor UAV was established by using Newton-Euler method.On the basis of the dynamic model,the control system framework was designed,which was composed of two closed-loop controls: inside-loop attitude control and outside-loop position control.Secondly,in attitude control,the back-stepping ADRC algorithm was formed by combining ADRC algorithm and back-stepping method.The comparison results was shown that the back-stepping ADRC algorithm had better control accuracy and anti-jamming performance than other control algorithms.In practice,the experimental platform of UAV attitude was built to verify that ADRC has better control effect than PID in attitude control.Then,because of the ADRC algorithm required many parameters,although the control effect was better when combined with the back-stepping method,it did not solve the problem of parameter tuning of ADRC algorithm.In order to select better parameters,this paper improved the PSO optimization algorithm according to the characteristics of the ADRC parameters,a two-space PSO optimization algorithm was formed into tuning the ADRC parameters.The simulation results shown that the two-space PSO optimization algorithm improved the optimization speed.The optimized parameters control effect was obviously better than the parameters before optimization.Finally,in position control,the commonly used PD control had incompatible parameters between fixed-point cruise and trajectory tracking.In order to solve this problem,a transition process was introduced to PD control,which made the parameters of the two modes consistent and improved the control performance.ADRC was adapted to improve the control effect,but considering the control effect and the hardware performance limitations,it was recommended to using the PD control with the transition process in position controls.